Computer Methods for Describing Groups from Binary Phenetic Data: Modification of Numerical Taxonomy Programs to Increase Flexibility

The logic of modifications to numerical taxonomy programs is described. The programs modified are to calculate and store interstrain similarities, to arrange the data by cluster analysis, and to calculate inter- and intragroup statistics. The modifications included increasing the efficiency of the similarity calculations, storing intermediate matrices on disks for later use, and permitting the calculation of group statistics on any arbitrary groups of strains as well as taxonomic groups without recalculating interstrain similarities. Changing communication paths of information allowed using the programs flexibly for ecological as well as taxonomic analyses, reduced program execution costs, increased program versatility, and reduced errors.